OR31

The global rise in Mycobacterium tuberculosis (Mtb) antimicrobial resistance means we urgently need new approaches to understand how it develops and stops its spread. One of the most common approaches used to counter the evolution of resistance in Mtb is the application of combination antibiotic therapy. Evolving resistance to combinations of drugs should be extremely rare, as it requires multiple mutations to occur in the same genetic background before microbial growth is inhibited. While wild-type bacteria cannot achieve this, mutators (with defects in DNA repair genes) allow multi-drug resistance to evolve easily during single-drug and combination treatments when there is a delay in reaching inhibitory concentrations of antibiotics. Therefore, it is important to identify these mutators to predict how combination therapies will be effective in preventing resistance. Here, our goal was to identify mutators contributing to resistance beyond known resistance genes. We first used a comprehensive bioinformatics approach where we analysed 53589 whole-genome sequences of Mtb clinical isolates, then we performed an association analysis and identified 151 mutator candidates that are associated with drug resistance in all major DNA repair pathways. Next, we used tools for Computational Biology to select different candidates for the experimentation. We finally constructed mutants by introducing point mutations on the chromosome of M. smegmatis a model species for Mtb and validated the hyper-mutability of five mutants using frequency tests conferring Rifampicin and Isoniazid. Work to extend these experiments to Mtb is underway to unveil new strategies to predict and combat the development of drug resistance.